ChatGPT/Dall-E generates images like nothing else. I'm a writer and have some complex characters, including a set of twin brothers that despite being identical, present themselves differently. No kidding, ChatGPT actually remembers their information and generates consistent characters...most of the time.
This is exactly what im looking to do with chat gpt. I want to write a series of children's books with the 2 main characters modelled after my own kids likeness. Is it good for this?
@@peterzammit9029 Yes, you can do this but they likely won't be exact replicas of your kids. To get the best results, use high-quality 1 front and 2 side photos.
The best explanation of ChatGPT using DALL E 3 I've come so far! Excellent presentation and helpful for a beginner who has just started to learn about this. Kudos to you and your team!
I've found putting a space between the letters when wanting text, helps ..most of the time.✌🏽(like for the 'i' or 'I' - it may not "see it" so spacing between can help.)
Dall-E is something I dont seem to get it work. I want a human-like work, for example, designing a web page or drawing an icon or making a color background etc.. Dall-E so artificial I give up.
Wow! I'm really impressed with the images you generated, Ronny! I especially loved the moodboards. I hadn't seen anything like that generated by AI before. And when you dissed Midjourney? That cracked. Me. Up. 😂 Ever consider doing stand-up comedy, Ronny?
@@TeamRonDiWell THAT'll be a sad day. I LIVE for your and Diana's videos. (What d'ya mean "fangirl"? 🤔) But I'd LOVE to see you do stand-up too! Please give us all a heads up before you go on tour, so we can all buy good seats close to the action!
Hello! I MUCH MUCH MUCH prefer Dall e 2 to Dall e 3. Is there a prompt I can include in Dall e 3 that will cause it to provide Dall e 2 results? Thanks.
@@TeamRonDi Thank you! Canva and MS Design and Adobe Express are using some sort of Dall e 2....but I guess they're adding some sort of "realistic" filter that I guess most people prefer. I don't know how to modify my prompt to ask it to loose the filter. Bummer that ChatGPT is so stupid :(
Yes, I've noticed that change. I guess it's just OpenAI being a bit less generous with image generation. The DALL-E GPT gives us two images at the moment.
Hey there Alexksandra, Like I explain in the beginning of the video, you need to have a ChatGPT Plus account to use it in ChatGPT. Or, you can use DALL-E 3 for free with the Bing image creator: www.bing.com/create
DALL-E 3 comes for all ChatGPT Plus users. Look under "Explore GPTs" and then scroll down to the "By ChatGPT" section. DALL-E should be there. The DALL-E you see is DALL-E 3.
Thank you so much! I was confused to why mine is only Dall-E and not 3. When I asked chatgpt if it’s using Dall-E-3. It says no and that I need to go through Open AI website😂
Ronny, I have a question regarding adding a link to an image. Currently, you can add a link to an image, but it only works if you download it as a PDF. Are there any plans to expand that so I can download the image as a PNG or Jpeg with a link so as to make the image clickable and allow me to post it in social media as a clickable image? Thanks
That is not possible unfortunately. For two reasons: PNGs or JPG do not support links, and second, social media platforms don’t want you to send traffic away from their platforms 😊
Right, It’s been quite aleatory TBH. Sometimes I got up to 4, but I believe OpenAI is defaulting to 1 nowadays to save on network usage. You can try to force it to give you 2 or 4 using Custom Instructions or even by asking it in your prompt.
If you are after high quality, I’d generate the stickers individually, instead of in the form of a set. This should help the model create more perfect outcomes. Also focus on common animals or objects (for which there are lots of images on the Internet). Remember that the models have been trained on these data sets that were available online.
Hi Guys, I'm confused! I clicked join, and followed the payment sign up. Did that so I seem to be a member. Then I go to the membership tab and find that I seem to be a member without the benefits suggested in this video??? It seems I'm in a teir that has been closed??? can you advise please? Thanks
Hey there 👋 We have two tiers in the membership, tier 02 (called “🚀 The RonDi-Ship” is the one with the perks explained in the video. It seems you figured it out 👏 Welcome aboard, you can watch our member updates to understand how to access our perks: th-cam.com/play/PL2jU5A3fxwC8Unq80MMw3_iELZ5d8drDM.html&si=Y-7exZm1PfWdkf-z
Hi dear, this is unusual 🤔 Here's what I would do: - Refresh your browser - Try a different browser - restart your computer IF none of that works, please contact the Udemy Support, as there's not much I can do from my (instructor) side. I know the course works since over 200 students are taking it right now. I hope you solve this issue quickly and start enjoying the course 😊🙏
@@TeamRonDi Ty for always replying. You are right! I tried a different browser. I started applying what I have learned in your course Dall E while watching. Im only half way in the course & Im so surprised on the outcome. Your genius is making my life easier! You're a blessing! hooray to team RonDi!
Hi there, I would like to know if is possible talk to you about making a Canva course in romanian language. there is a way to speak with you ? Thank you so much
Ah Dalle-3 is not the easiest. Dalle revises every prompt so you never can get control on the output like with Midjourney. where parameters have effect (and now the new cref option to reference to an image to get even more control over the output.). Dalle-3 generates nice things. i have used the openai API for months now and have gotten very nice photos , images and the weirdest artlike output that i never could have prompted. i guess for now Dalle-3 is like Salvador Dale himself ,able to create unimaginable artlike output ,where Midjourney is generating Photorealistic output. In the end the output must have a commercal added value and so i would use Dalle-3 for ArtyFarty output and Midjournery for Photos and Discord: that works actually very well for the new generation. but that is another topic for another discussion. regarding adding text that must be inserted into the image: this was not in scope of Dalle-3. maybe that was fixed?
Thanks for the detailed feedback dear I have to admit, I’ve only used Midjourney a couple of times. It’s probably really good too. The discord thing is just extra friction for me.
ok i know dalle 3 is not the strongest part of chatgpt but anyway though i just now watched a video in tips in mj and the images there are just.. superb. these are quite meh in comparison which is confusing.. why dalle 3 team hasnt moved forward yet that much
💓Go deeper with our DALL-E Master Course 👉bit.ly/DALL-EMasterCourse
Great video. Very engaging. Did you outsource the animations and titles in the video or do them yourself? If so, which software did you use?
😂
ChatGPT/Dall-E generates images like nothing else. I'm a writer and have some complex characters, including a set of twin brothers that despite being identical, present themselves differently. No kidding, ChatGPT actually remembers their information and generates consistent characters...most of the time.
I also love using DALL-E and hope it will keep improving!
This is exactly what im looking to do with chat gpt. I want to write a series of children's books with the 2 main characters modelled after my own kids likeness. Is it good for this?
@@peterzammit9029 Yes, you can do this but they likely won't be exact replicas of your kids. To get the best results, use high-quality 1 front and 2 side photos.
The best explanation of ChatGPT using DALL E 3 I've come so far! Excellent presentation and helpful for a beginner who has just started to learn about this. Kudos to you and your team!
Great to hear you found the tutorial helpful 💜✨
Watch out for our DALL-E course coming soon!
The way it demonstrates how to create diverse images with just a few prompts in ChatGPT using DALL·E 3 blew my mind.
Yes!
That is the main reason why I love using DALL-E in ChatGPT. It helping us create more elaborate prompts ✨
The only problem is it makes a lot of spelling mistakes in the image which makes it difficult to use in any project.
I've got a bit of an advantage with the misspelling. I'm an artist who works with Krita, so I can make spelling corrections when working with DALL-E.
I've found putting a space between the letters when wanting text, helps ..most of the time.✌🏽(like for the 'i' or 'I' - it may not "see it" so spacing between can help.)
another super helpful tutorial! always learning from y'all x)
Yay, thank you!
Dall-E is something I dont seem to get it work. I want a human-like work, for example, designing a web page or drawing an icon or making a color background etc.. Dall-E so artificial I give up.
Definitely we have a different vision of what “hyper-realistic” is. All the images have the same illustration look to me.
Great, informative video and inspired me for an afternoon of designing. I needed that!
Glad I could inspire you and get you to play with DALL-E!
Great explanation brother, helped me a lot. Love from India ❤
Glad it helped 💜
I find this super insightful. Ty for the information abt Dall.E 3! Your team rocks😊
Yayy 🙌
I'm happy to hear you enjoyed the tutorial!
Much more to come on the channel about DALL-E, for sure
Could you please share the colour codes you used in thumbnail?
Do we get unlitted access to DALL-E on getting GPT plus? Or is it that we have to buy the DALL-E credits too? Can someone tell, please?
There's still a daily limit of images you can generate. But you don't need to buy credits, just wait a few hours.
@@TeamRonDi thanks 👍
Wow! I'm really impressed with the images you generated, Ronny! I especially loved the moodboards. I hadn't seen anything like that generated by AI before.
And when you dissed Midjourney? That cracked. Me. Up. 😂
Ever consider doing stand-up comedy, Ronny?
Haha, I love stand up comedy!
Maybe one day when I retire from TH-cam 😊
@@TeamRonDiWell THAT'll be a sad day. I LIVE for your and Diana's videos. (What d'ya mean "fangirl"? 🤔)
But I'd LOVE to see you do stand-up too! Please give us all a heads up before you go on tour, so we can all buy good seats close to the action!
@@TeamRonDi Do you think DALL-E 3 offers equal image quality now to MJ?
Can we sell this images on any stock website?
If they are made with DALL-E, yes.
Thanks for the tips! I would love to know how you made the short clips with your image and the prompts? Did you make them in Canva Pro?
Yes, we create these slides in Canva and use the Canva animations 😊
I just joined your channel at the Jalapeno level and can't find these prompts anywhere. Please link if possible :)
Hi there 👋
Prompts, courses and Live sessions are Tier 2 perks (🚀 The RonDi-Ship)
Does the paid version come with unlimited generations?
No, there will still be a limit to the amount of images you can generate.
Do we have to get DALL-E credits too in order to generate images or GPT plus alone is enough to create the limited number of images?
Thank you for great video Rony
My pleasure 😇
Hello! I MUCH MUCH MUCH prefer Dall e 2 to Dall e 3. Is there a prompt I can include in Dall e 3 that will cause it to provide Dall e 2 results? Thanks.
Not that I know of, unfortunately. Canva still uses DALL-E 2 via it’s DALL-E app.
@@TeamRonDi Thank you! Canva and MS Design and Adobe Express are using some sort of Dall e 2....but I guess they're adding some sort of "realistic" filter that I guess most people prefer. I don't know how to modify my prompt to ask it to loose the filter. Bummer that ChatGPT is so stupid :(
Any idea why Chat GPT only renders one image per prompt now as opposed to four?
Yes, I've noticed that change.
I guess it's just OpenAI being a bit less generous with image generation.
The DALL-E GPT gives us two images at the moment.
Which software you are using for captions
Our editors create them in Premiere Pro.
Can i modify the images from dall-e on canva?
Yes, it will be considered as a photo, so you’ll have access to all the photo effects and filters.
Thank you showing this.
My pleasure
Another awesome video! But this option is not free or am I wrong?
Hey there Alexksandra,
Like I explain in the beginning of the video, you need to have a ChatGPT Plus account to use it in ChatGPT.
Or, you can use DALL-E 3 for free with the Bing image creator: www.bing.com/create
Thanks a lot, I missed the info🙆♀, I'll try! 👍@@TeamRonDi
Is it possible to give ChatGPT an element to include in a design?
Not that I know of. You'll have to describe that element in your prompt.
I have chatgpt plus and don’t have Dall-E-3. only Dall-E and the images are not realistic. How do I enable Dall-E-3?
DALL-E 3 comes for all ChatGPT Plus users.
Look under "Explore GPTs" and then scroll down to the "By ChatGPT" section. DALL-E should be there. The DALL-E you see is DALL-E 3.
Thank you so much! I was confused to why mine is only Dall-E and not 3. When I asked chatgpt if it’s using Dall-E-3. It says no and that I need to go through Open AI website😂
Ronny, I have a question regarding adding a link to an image. Currently, you can add a link to an image, but it only works if you download it as a PDF. Are there any plans to expand that so I can download the image as a PNG or Jpeg with a link so as to make the image clickable and allow me to post it in social media as a clickable image? Thanks
That is not possible unfortunately.
For two reasons:
PNGs or JPG do not support links, and second, social media platforms don’t want you to send traffic away from their platforms 😊
How do you get multiple image results? I only get 1 result when I try.
Right,
It’s been quite aleatory TBH.
Sometimes I got up to 4, but I believe OpenAI is defaulting to 1 nowadays to save on network usage.
You can try to force it to give you 2 or 4 using Custom Instructions or even by asking it in your prompt.
Thanks Ronnie!❤
You are welcome dear 😊
hi, can we generate very good quality sticker designs with it? please answer me
If you are after high quality, I’d generate the stickers individually, instead of in the form of a set. This should help the model create more perfect outcomes. Also focus on common animals or objects (for which there are lots of images on the Internet). Remember that the models have been trained on these data sets that were available online.
Hi Guys, I'm confused! I clicked join, and followed the payment sign up. Did that so I seem to be a member. Then I go to the membership tab and find that I seem to be a member without the benefits suggested in this video??? It seems I'm in a teir that has been closed??? can you advise please? Thanks
Hi again. Sorry I went back and realised I needed to 'upgrade' I've done that. so problem solved. DOH! LOL Thanks
Hey there 👋
We have two tiers in the membership, tier 02 (called “🚀 The RonDi-Ship” is the one with the perks explained in the video.
It seems you figured it out 👏
Welcome aboard, you can watch our member updates to understand how to access our perks: th-cam.com/play/PL2jU5A3fxwC8Unq80MMw3_iELZ5d8drDM.html&si=Y-7exZm1PfWdkf-z
It doesn't allow me to use my own image of myself and add it to an overall scene. Anyway to do that?
That’s because of its safety filters.
I’ll try to find a way around that.
Thank you. What about the API?
We haven't yet looked into the DALL-E API ... We'll create some content about it when we do 😊
Hi teacher Ronnie! I bought your Dall-e course in Udemy but it doesn't play and I paid. pls help me
Hi dear,
this is unusual 🤔
Here's what I would do:
- Refresh your browser
- Try a different browser
- restart your computer
IF none of that works, please contact the Udemy Support, as there's not much I can do from my (instructor) side. I know the course works since over 200 students are taking it right now.
I hope you solve this issue quickly and start enjoying the course 😊🙏
@@TeamRonDi Ty for always replying. You are right! I tried a different browser. I started applying what I have learned in your course Dall E while watching. Im only half way in the course & Im so surprised on the outcome. Your genius is making my life easier! You're a blessing! hooray to team RonDi!
Isn’t this available in canva? ❤️🇦🇺
In Canva you get DALL-E 2 (via Apps). Which is significantly less powerful.
How many images i can create per day in chatgpt 4 dall e 3..??
From what I've seen recently, you can prompt 40 times per 3h.
Hi there, I would like to know if is possible talk to you about making a Canva course in romanian language. there is a way to speak with you ? Thank you so much
Hi dear,
We have a business enquiries email in our videos, you can reach out there and tell us a bit more about your project 😊
thank you so much ! Have a good day. And Ihope to hear you soon .@@TeamRonDi
Thank you for this video.
Happy to help, Jackie
How come some vloggers says that Dalle-3 is free...???😢😢😢
You can use DALL-E for free via the Microsoft Copilot or Bing.
Just google: Bing Image Creator.
How to generate a picture of Dwayne Arnold?
I don't think DALL-E will let you generate images of known people 🥺
@@TeamRonDi do you know this channel " Dwayne, Arnold, Neil deGrasse Tyson" youtube.com/@reallynotai?si=G4J1WQouP07990eP
So, if I have a ChatGPT pro subscription for $20/month, that means I can generated unlimited HD images with DALL-E 3?
Not unlimited, but yes you have access to DALL-E 3 with a GPT Plus account.
@@TeamRonDiThank you, how is it not unlimited?
Hi. I'm your first commenter
And we you for that!
i woder if we can use this Ai in Canvas
You can access DALL-E 2 via the DALL-E app in Canva, not FALL-E 3 unfortunately 😢
Fantastic
Thank you! Cheers!
Ah Dalle-3 is not the easiest. Dalle revises every prompt so you never can get control on the output like with Midjourney. where parameters have effect (and now the new cref option to reference to an image to get even more control over the output.). Dalle-3 generates nice things. i have used the openai API for months now and have gotten very nice photos , images and the weirdest artlike output that i never could have prompted.
i guess for now Dalle-3 is like Salvador Dale himself ,able to create unimaginable artlike output ,where Midjourney is generating Photorealistic output. In the end the output must have a commercal added value and so i would use Dalle-3 for ArtyFarty output and Midjournery for Photos
and Discord: that works actually very well for the new generation. but that is another topic for another discussion.
regarding adding text that must be inserted into the image: this was not in scope of Dalle-3. maybe that was fixed?
Thanks for the detailed feedback dear
I have to admit, I’ve only used Midjourney a couple of times. It’s probably really good too. The discord thing is just extra friction for me.
Hi Will Dall-3 use my idea without my permission
You can refer to Open AI's policy on usage of your data here: openai.com/policies/usage-policies/
ok i know dalle 3 is not the strongest part of chatgpt but anyway though i just now watched a video in tips in mj and the images there are just.. superb. these are quite meh in comparison which is confusing.. why dalle 3 team hasnt moved forward yet that much
Cooool
Sucks, you can't upload an image to change it.
That’s not what it’s intended for 🤓
Too bad about the text
Yes!
Though I'm sure this issue will be fixed in all Text to Image tools soon.
JESUS❤
TH-camrs are full of lies, no one has made any video of dalle 3 turning images upside down
Not sure what you mean 🤔
It’s too slow and too limited for $20 a month. Become familiar with the open source models and loras it’s way more flexible and free.
Thanks for sharing 😊
Could you please share the colour codes you used in thumbnail?
These are our brand colors 😊
We’d suggest you create a different palette for your brand!